Drug addiction is a chronically relapsing disorder that often has devastating consequences for the addicted person and society as a whole. Emerging evidence suggests that one possible reason for this cycle of drug taking and relapse may center on cognitive impairments. Chronic exposure to drugs of abuse like cocaine can have deleterious effects on general learning and executive functions by disrupting neural encoding in limbic structures critical for cognitive processes such as the basolateral amygdala (BLA) and a target of BLA projections, the nucleus accumbens (NAc). Indeed, this same circuit is known to be critical for supporting both simple and more cognitive higher-order learning in normal (drug-nave) animals. Thus, understanding the normal neural processing in this circuit and how it is altered by repeated cocaine experience will be essential for understanding some of the critical components of addiction and provide potential therapeutic avenues for possible treatments. Here, I hypothesize that normal learning is characterized by the interaction of BLA input to NAc neurons which are modulated by dopaminergic (DAergic) inputs arising from the ventral tegmental area (VTA). Following repeated cocaine exposure, this circuit will be disrupted, characterized by poor BLA input to NAc and impoverished DA signaling. This dysfunction will prevent the normal encoding of task-relevant representations and consequently impair behavior. During the mentored K99 phase, I will thoroughly explore this functional circuit using complementary techniques. First, I will characterize DA signaling using fast-scan cyclic voltammetry (FCSV) in rats with and without a prior history of cocaine self-administration while performing a second-order Pavlovian learning task. Next, I will learn optogenetic techniques and apply them in a transgenic line of TH::Cre rats which will allow for specific targeting of DA neurons in VTA. With this technique, I will be able to manipulate DA release specific to the VTA-NAc pathway, allowing the ability to see whether brief pauses in DA release are sufficient to block learning in normal animals, and if transient bursts of DA are sufficient to rescue cognitive function. In my independent R00 phase, I will explore the contributions of the BLA to this system by characterizing BLA neural encoding with in vivo electrophysiology of the second- order task in rats with a history of cocaine experience or yoked controls. Finally, I will use data from that experiment to see whether optical inhibition of BLA afferents selective to the NAc is able to alter learning in normal animals, but restore function in rats with a history of cocaine self-administration. Taken together, this proposal will provide a thorough characterization of the BLA-NAc-VTA neural circuit after repeated cocaine exposure, using a higher-order learning task to parse specific cognitive deficits.